Commits · 726e12cf0ce6c926877bd8e19e6afe05e5101eeb · Raul Torres / llvm-target-spread · GitLab

Sep 19, 2018
- ScheduleDAG: Cleanup dumping code; NFC · 726e12cf
  Matthias Braun authored Sep 19, 2018
  726e12cf
Sep 18, 2018
- [AMDGPU] Match udot8 pattern · f5a28483
  Farhana Aleen authored Sep 18, 2018
  f5a28483
- AMDGPU: Don't form fmed3 if it will require materialization · ebf46143
  Matt Arsenault authored Sep 18, 2018
  ebf46143
- AMDGPU: Expand vector canonicalizes · 9d49c449
  Matt Arsenault authored Sep 18, 2018
  9d49c449
Sep 17, 2018
- [AMDGPU] Initialize instruction itinerary from GCNSubtarget · 06d3b413
  Stanislav Mekhanoshin authored Sep 17, 2018
  06d3b413
Sep 14, 2018
- AMDGPU: Clear the bits before they are being set in program resource registers · e721b11c
  Konstantin Zhuravlyov authored Sep 14, 2018
  e721b11c
- [AMDGPU] Ensure trig range reduction only used for subtargets that require it · 20de3e99
  David Stuttard authored Sep 14, 2018
  20de3e99
Sep 13, 2018
- [AMDGPU] Removed unused method · c8af6a46
  Tim Renouf authored Sep 13, 2018
  c8af6a46
- AMDGPU: Fix not preserving alignent in call setups · ff987ac6
  Matt Arsenault authored Sep 13, 2018
  ff987ac6
- [AMDGPU] Load divergence predicate refactoring · 4d302f69
  Alexander Timofeev authored Sep 13, 2018
  4d302f69
- [AMDGPU] Preliminary patch for divergence driven instruction selection.... · 2fb44808
  Alexander Timofeev authored Sep 13, 2018
  2fb44808
Sep 12, 2018
- AMDGPU: Print all kernel descriptor directives (including the ones with default values) · 6e551e0e
  Konstantin Zhuravlyov authored Sep 12, 2018
  6e551e0e
- AMDGPU: Re-apply r341982 after fixing the layering issue · 71e43ee4
  Konstantin Zhuravlyov authored Sep 12, 2018
  71e43ee4
- Revert "AMDGPU: Move isa version and EF_AMDGPU_MACH_* determination into TargetParser." · 95066496
  Ilya Biryukov authored Sep 12, 2018
  95066496
Sep 11, 2018
- AMDGPU: Move isa version and EF_AMDGPU_MACH_* determination · 941615e4
  Konstantin Zhuravlyov authored Sep 11, 2018
  941615e4
- [AMDGPU] Preliminary patch for divergence driven instruction selection.... · db7ee766
  Alexander Timofeev authored Sep 11, 2018
  db7ee766
- AMDGPU: Remove leftovers from configurable address spaces · 796b0e7a
  Matt Arsenault authored Sep 11, 2018
  796b0e7a
Sep 10, 2018
- [AMDGPU] Preliminary patch for divergence driven instruction selection.... · 20cbe6f3
  Alexander Timofeev authored Sep 10, 2018
  20cbe6f3
- AMDGPU: Remove function pointer type hack · d1f4571a
  Matt Arsenault authored Sep 10, 2018
  d1f4571a
- AMDGPU: Stop reporting is-noop addrspacecast for constant 32-bit · 7f6dc597
  Matt Arsenault authored Sep 10, 2018
  7f6dc597
- DAG: Handle odd vector sizes in calling conv splitting · 57b5966d
  Matt Arsenault authored Sep 10, 2018
  57b5966d
- [AMDGPU] Prevent sequences of non-instructions disrupting GCNHazardRecognizer wait state counting · f898edd1
  Carl Ritson authored Sep 10, 2018
  f898edd1
- AMDGPU: Use GOT PSV since it has an address space now · d77fcc2a
  Matt Arsenault authored Sep 10, 2018
  d77fcc2a
- AMDGPU: Don't abort on unknown addrspace argument · b9986746
  Matt Arsenault authored Sep 10, 2018
  b9986746
Sep 07, 2018
- [AMDGPU] Preliminary patch for divergence driven instruction selection. Fold immediate SMRD offset. · a805c96c
  Alexander Timofeev authored Sep 07, 2018
  a805c96c
Sep 06, 2018
- Revert r341413 · 834cbc64
  Scott Linder authored Sep 06, 2018
  834cbc64
- AMDGPU: Remove old hack for function addresses · df84dc69
  Matt Arsenault authored Sep 06, 2018
  df84dc69
Sep 04, 2018
- [AMDGPU] Legalize VGPR Rsrc operands for MUBUF instructions · dfe089df
  Scott Linder authored Sep 04, 2018
  dfe089df
- AMDGPU: Fix DAG divergence not reporting flat loads · 813613c4
  Matt Arsenault authored Sep 04, 2018
  813613c4
Sep 03, 2018
- Remove unnecessary semicolon to silence -Wpedantic warning. NFCI. · 2e35c1e3
  Simon Pilgrim authored Sep 03, 2018
  2e35c1e3
Sep 01, 2018
- AMDGPU/GlobalISel: Define instruction mapping for G_SELECT · ffc6bd6f
  Tom Stellard authored Sep 01, 2018
  ffc6bd6f
- [AMDGPU] Split v32i32 loads · 44451b33
  Stanislav Mekhanoshin authored Aug 31, 2018
  44451b33
Aug 31, 2018
- AMDGPU: Restrict extract_vector_elt combine to loads · bf07a50a
  Matt Arsenault authored Aug 31, 2018
  bf07a50a
- AMDGPU: Stop forcing internalize at -O0 · 988df635
  Matt Arsenault authored Aug 31, 2018
  988df635
- AMDGPU: Remove remnants of old address space mapping · 0da6350d
  Matt Arsenault authored Aug 31, 2018
  0da6350d
Aug 30, 2018
- [NFC] Rename the DivergenceAnalysis to LegacyDivergenceAnalysis · 35617ed4
  Nicolai Haehnle authored Aug 30, 2018
  35617ed4
- [AMDGPU] Preliminary patch for divergence driven instruction selection. Operands Folding 1. · 201f892b
  Alexander Timofeev authored Aug 30, 2018
  201f892b
Aug 29, 2018
- AMDGPU: Handle 32-bit address wraparounds for SMRD opcodes · 3fc2079c
  Marek Olsak authored Aug 29, 2018
  3fc2079c
- [AMDGPU] Match udot4 pattern. · 9250c92d
  Farhana Aleen authored Aug 29, 2018
  9250c92d
- AMDGPU: Fix getInstSizeInBytes · 283b9950
  Nicolai Haehnle authored Aug 29, 2018
  283b9950